26 lines
746 B
Python
26 lines
746 B
Python
import struct
|
|
import base64
|
|
import logging
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
def float32_to_s16le(float32_bytes: bytes) -> bytes:
|
|
num_samples = len(float32_bytes) // 4
|
|
floats = struct.unpack(f"<{num_samples}f", float32_bytes)
|
|
int16_samples = [max(-32768, min(32767, int(f * 32767.0))) for f in floats]
|
|
return struct.pack(f"<{num_samples}h", *int16_samples)
|
|
|
|
|
|
def build_display_text(accumulated: str, current: str) -> str:
|
|
parts = [p for p in (accumulated, current) if p and p.strip()]
|
|
return " ".join(parts)
|
|
|
|
|
|
class ASRClient:
|
|
def __init__(self, settings):
|
|
self.settings = settings
|
|
|
|
async def transcribe_full(self, audio_bytes: bytes, language: str = "yue") -> str:
|
|
raise NotImplementedError
|